Select to expand quote

jn1 said..
Did you try interchanging known good BN220 unit with a bad unit, then repeating the test a few times ?, and then comparing the data ?

Yes, of course I verified that it's not the logger. Here's a picture of another check:

The red logger on the left is new, the green and white loggers are old and proven to work. I switched the GPS chip on the green logger to one of the new BN220s. The picture is after 15+ minutes on the front lawn with a clear sky view. The new BN220s get only about half as many sats as the Beitian BN220.

I then turned all units off, switched the green unit back to the original Beitian chip, and put the new BN220 on the other old and proven logger. Picture after about 7 minutes:


Again, both new BN220s get only half the satellites compared to the old one. Note that this was a hot start for the left and right units, but a warm start for the green unit. It still picked up 23 sats within less than 10 minutes, which is exactly the expected behavior. The new chips never got higher than 14. They are defective. QED.

Select to expand quote

boardsurfr said..

jn1 said..
Did you try interchanging known good BN220 unit with a bad unit, then repeating the test a few times ?, and then comparing the data ?

Yes, of course I verified that it's not the logger. Here's a picture of another check:

The red logger on the left is new, the green and white loggers are old and proven to work. I switched the GPS chip on the green logger to one of the new BN220s. The picture is after 15+ minutes on the front lawn with a clear sky view. The new BN220s get only about half as many sats as the Beitian BN220.

I then turned all units off, switched the green unit back to the original Beitian chip, and put the new BN220 on the other old and proven logger. Picture after about 7 minutes:


Again, both new BN220s get only half the satellites compared to the old one. Note that this was a hot start for the left and right units, but a warm start for the green unit. It still picked up 23 sats within less than 10 minutes, which is exactly the expected behavior. The new chips never got higher than 14. They are defective. QED.

are they all Beitian branded, or just generic BN 220

Select to expand quote

boardsurfr said..

jn1 said..
Did you try interchanging known good BN220 unit with a bad unit, then repeating the test a few times ?, and then comparing the data ?

Yes, of course I verified that it's not the logger. Here's a picture of another check:

The red logger on the left is new, the green and white loggers are old and proven to work. I switched the GPS chip on the green logger to one of the new BN220s. The picture is after 15+ minutes on the front lawn with a clear sky view. The new BN220s get only about half as many sats as the Beitian BN220.

I then turned all units off, switched the green unit back to the original Beitian chip, and put the new BN220 on the other old and proven logger. Picture after about 7 minutes:


Again, both new BN220s get only half the satellites compared to the old one. Note that this was a hot start for the left and right units, but a warm start for the green unit. It still picked up 23 sats within less than 10 minutes, which is exactly the expected behavior. The new chips never got higher than 14. They are defective. QED.

Just out of interest, what's the voltages across the power rails of a defective BN220 compared to a new BN220 on the white unit ?

Select to expand quote

Windxtasy said..
are they all Beitian branded, or just generic BN 220

All working units are Beitian branded. Over time, I got them from at least 4 different suppliers. The ones that are not working are generic BN220s, all from the same supplier and order. I tested 5 of the 6 I received, they all show the same problem.

When previously ordering from Amazon, I had noticed that they offered generic BN220s at a lower price than the Beitian BN220s, but I shied away from them. I probably just searched for "BN220" (without "Beitian") when placing the order on AliExpress. I did, however, check the specifications, and they are virtually identical.

Select to expand quote

jn1 said..
Just out of interest, what's the voltages across the power rails of a defective BN220 compared to a new BN220 on the white unit ?

The new BN220 on the white unit is the one that is defective, so your question makes no sense.

I have wasted quite a few hours just making sure that the new BN220 chips are indeed not up to snuff. I won't spend any more time trying to figure out exactly why. I'll simply stick to Beitian BN-220 chips from now on, and order from a supplier where returns are not a problem. Wasting hours trying to save a few bucks makes no sense. AliExpress has offered a refund of $15 on the $78 order, which is ridiculous.

Select to expand quote

boardsurfr said..

Windxtasy said..
are they all Beitian branded, or just generic BN 220

All working units are Beitian branded. Over time, I got them from at least 4 different suppliers. The ones that are not working are generic BN220s, all from the same supplier and order. I tested 5 of the 6 I received, they all show the same problem.

When previously ordering from Amazon, I had noticed that they offered generic BN220s at a lower price than the Beitian BN220s, but I shied away from them. I probably just searched for "BN220" (without "Beitian") when placing the order on AliExpress. I did, however, check the specifications, and they are virtually identical.

Mine are Beitian branded BN220s and I still only get 14 - 16 sats

Select to expand quote

Windxtasy said..
Mine are Beitian branded BN220s and I still only get 14 - 16 sats

Which firmware version are you using? 14-16 sats is normal for older firmware that used only 2 GNSS systems.

You need to (a) use a current firmware version, and (b) modify the config files for each unit by adding an entry to use 3 GNSS systems, for best performance. You should be able to routinely get more than 20 satellites while windsurfing.

After updating the firmware and changing the config file, check the screens at startup to make sure it shows "config OK". Then, leave the units out with a clear, unobstructed view of the sky for at least 30 minutes so that they can get all the info about the Galileo satellites. That may require only 10-15 minutes if you are lucky (just check the number of satellites reported).

Select to expand quote

boardsurfr said..

Windxtasy said..
Mine are Beitian branded BN220s and I still only get 14 - 16 sats

Which firmware version are you using? 14-16 sats is normal for older firmware that used only 2 GNSS systems.

You need to (a) use a current firmware version, and (b) modify the config files for each unit by adding an entry to use 3 GNSS systems, for best performance. You should be able to routinely get more than 20 satellites while windsurfing.

After updating the firmware and changing the config file, check the screens at startup to make sure it shows "config OK". Then, leave the units out with a clear, unobstructed view of the sky for at least 30 minutes so that they can get all the info about the Galileo satellites. That may require only 10-15 minutes if you are lucky (just check the number of satellites reported).

I am using the older firmware because I don't want all the extra screens and things that have been added since, but I would like to be able to use all the satellites.

So you'll need to upgrade. 14-16 satellites is just borderline usable, and the quality of your data will go up if you enable a third GNSS, which you can only do in the newer firmware. As far as I understand, the new screens are all additional, so you should still see the same screens while surfing after the upgrade.

I've built my 1st unit. Thanks Jan for your idea . I followed Jan's exact recipe, accept I made my own IP67 power connector (1st cut design). I did a few stupid stuff ups with this 1st build, but all learning.

Nb/ Regarding the GOPRO Hero 10 enclosures, don't bother drilling holes through the scratch proof lens cover. Fark me, I burnt out a 3mm drill bit trying to tap a pilot hole through it (this is where I was going to fit my power connector, and why socket neck is a bit too short where it is).

The connector has an in series Schottky power diode. The fasteners are Nylon M10. The wires are 18 AWG copper. The charger cable is a recycled USB A-B printer cable with a two contact header connector. So, unit charges from USB power.

Select to expand quote

boardsurfr said..
So you'll need to upgrade. 14-16 satellites is just borderline usable, and the quality of your data will go up if you enable a third GNSS, which you can only do in the newer firmware. As far as I understand, the new screens are all additional, so you should still see the same screens while surfing after the upgrade.

I installed the newest software and got 27 satellites on both units for today's session. All the screens are the same except for the sleep screen

Epoxied neck on my home made socket. Connector protector as cover. Works well.


My 13mm diameter connectors arrived. So, I'll use them for my next builds.

www.amazon.com.au/SZJELEN-Electronics-Connector-Outdoor-Waterproof/dp/B07C4KQLPD

I have managed to build 2 units and loaded the software. the units switch on and start recording data as expected but once I end the gps recording by activating the reed switch or IO39 the units go to sleep and I am unable to get them to restart. I have tried toggling the on off switch, nothing works. I have to reinstall the software and then it works again. Only to stop work again when switched off. Can anyone suggest a solution?

Thanks

This is a rather strange behaviour. Can you measure the voltages between GPIO39 and GND, Reset and GND ? They should have 3.3 V. When pressing GPIO39 (or reed), GPIO39 should go to 0V. When pressing RST, RST should also go to 0V. Vbat should always show the battery voltage (3V - 4.2 V).
Another test : if you load the sw with Arduino, try to read the messages with the Terminal window. If you loaded the sw with the ESP-tool, you first need a terminal program (i use Hterm : www.der-hammer.info/pages/terminal.html) to read the messages.

I measured the voltage on 39 and RST to GND and I was only getting 0.7V
I disconnected the battery and checked the battery voltage which was 3.7V then reconnected the battery to the board.
When I test 39 and RST to GND now I get 3.7V but the unit will still not switch back on?

The 3.7V on RST and GPIO39 is too high, should be 3.3V . Maybe your voltageregulator on the board is defect. If you power the board over USB, do you have the the same problems ? Can you connect the battery directly to the solderpad "VBAT" ? (there is a fuse on the bat connector which is prone to fail).

Select to expand quote

chaseboost said..
I have managed to build 2 units and loaded the software. the units switch on and start recording data as expected but once I end the gps recording by activating the reed switch or IO39 the units go to sleep and I am unable to get them to restart. I have tried toggling the on off switch, nothing works. I have to reinstall the software and then it works again. Only to stop work again when switched off. Can anyone suggest a solution?

Thanks

I had one unit that showed the same behavior. It froze at the screen before the final sleep screen, and did not wake up again, even after disconnecting power. This did not happen right away after building, everything was fine for a few stationary tests and one or two tests on the water.

For a few days, it eventually switched back on after being fully powered down for half an hour or longer. But eventually, it did not turn on again at all (Edit: I just plugged the chip in again after it had been lying around for a week or two, and it came back to life; still freezes when pressing the reset button, and remains catatonic then).I tried removing the 39 switch from the board, but that did not change anything. I figured it was a bad board, since the other 9 units I have built all worked fine.

Select to expand quote

rp6conrad said..
The 3.7V on RST and GPIO39 is too high, should be 3.3V . Maybe your voltageregulator on the board is defect. If you power the board over USB, do you have the the same problems ? Can you connect the battery directly to the solderpad "VBAT" ? (there is a fuse on the bat connector which is prone to fail).

On my ESP32 chip that shows very similar behavior to what chaseboost reported, the problem looked the same with USB power.

I got a bit more information about what's going on with my bad chip. The "nothing happens when restarting" problem also happens when I use the power button on the chip. When connected to a USB power meter, the chip shows a use of 0.01A when switched on, so something is going on.
When hooked up to Arduino, the serial monitor shows the following error message:

rst:0x10 (RTCWDT_RTC_RESET),boot:0x17 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1216
ho 0 tail 12 room 4
load:0x40078000,len:10944
load:0x40080400,len:6388
csum err:0x04!=0xa6
ets_main.c 384
ets Jul 29 2019 12:21:46

From a brief Google search, this seems to be a problem somewhere in the flash chip that prevents the secure bootloader from executing when it detects the checksum error. Possible causes include voltage issues, EMI, and faulty chips. There are some references of ESP32 chips getting "bricked up" after various periods ranging from days to multiple months of use.

As suggested I connected the battery to VBAT and GND and both units worked as expected, inductive charging was working and all looked good however this evening I am getting the same unresponsive result when trying to power up.

Been trying for a while to get a reliable way to film the DIY GPS in action. Many wasted sessions trying to use the over complicated and unreliable GoPro Hero9 (GoPro however have been kind enough to send a new replacement even though out of warranty). Tried yesterday with the old but reliable Hero 5 and got this footage. Next issue was a mount that wouldn't vibrate too much or break. Tried aluminium but failed due to fatigue stress and lost the GPS. (Luckily found again next day). This mount seems more reliable made out of steel bolted to the rear tuttle with a Aluminium stabiliser mounted to the standard GoPro mount. The GPS filmed is my original epoxy'd unit still going strong after a year of abuse. The camera mount is 3D printed but strengthened with Al and then all leashed to the rear footstrap. Managed to get a PB Alpha in non optimal conditions thanks to Jan's GPS...hopefully this video shows how it works on the water

Great stuff Flex, only you could do something like that!
Did you have that on board when we talked yesterday?

Got my unit running since the early days on this thread. I am not a speed surfer, but interested in my progression on speed nonetheless, and love DIY projects. Gps-speedsurfing has some nice ways of keeping track of certain stats, I decided to create an account there, also paid for the subscription, at least to try it out.

All the stats seem to be logged fine from the unit, but I am not able to show the track nor the runs on the map. I Found that other users from this unit do have visualizations of their track. Is there something in the settings that I can change?

Did you already try to upload the .oao file ? It seems that with this file the map function is OK. Here the map from my last session on gps-speedsurfing.
www.gps-speedsurfing.com/default.aspx?mnu=user&val=392707&uid=23556

Yes Decrepit, all that stuff onboard yesterday when we talked. That's you standing mid first Alpha. Separate subject, can't paste images into GPSTC though like you...must be a Mac issue.

Select to expand quote

rp6conrad said..
Did you already try to upload the .oao file ? It seems that with this file the map function is OK. Here the map from my last session on gps-speedsurfing.
www.gps-speedsurfing.com/default.aspx?mnu=user&val=392707&uid=23556

Thanks for your reply.
I noticed that your sessions do indeed show the tracks and runs on the map. I uploaded an old .oao file that is on my hard drive. The session posted with the .oao does show the mapped runs and tracks. I will upload the oao files in the future.

Select to expand quote

Flex2 said..
Yes Decrepit, all that stuff onboard yesterday when we talked. That's you standing mid first Alpha. Separate subject, can't paste images into GPSTC though like you...must be a Mac issue.

Sorry I missed having a good look at it.
Sometimes copying the image, I just get the file path to it when I paste.
You have to make sure you're copying the visible image not the images path.
If it's in a format that GPSTC supports it should go in .jpg is what I mostly use. Also size may be a problem. I usually scale mine to 700 or 800 pixels wide

Select to expand quote

Flex2 said..
Been trying for a while to get a reliable way to film the DIY GPS in action.

Great video showing the alpha mode in action, Flex! Great turns, too. Pretty amazing alpha for a 27 knot top speed in the alpha run!

I've picked up 12 kits from Elmo, will start soldering them up sometime soon.
Anything interesting along the way, I'll report here.

OK first one wired up, and first fail!
Destroyed a reed switch. Put too much pressure on the glass where the wires exits, and the thing virtually fell to pieces.

Take home message, pre-bend the wires to fit the holes, holding wire between the bend and switch so no force is applied to the glass.

Took me a while but I'll get quicker.
I also think, the battery should be connected last, as it's the heaviest, and swinging on it's wires while soldering other stuff, doesn't seem ideal.

Plan is to try and get my two working first, as a learning exercise. So I'll wire another one up, then see if I can learn how to program them.